Neural Control of Movement Clear vision requires the ability to coordinate the movements of the eyes and head to control gaze (where we look in space). This project is devoted to understanding the nature of neuronal and muscular mechanisms required for clear vision. Our interest in normal behavior is refined by a focus on problems that lead to clinical eye movement disorders, such as ocular oscillations, misalignment of the two eyes (strabismus), and the vestibulo-ocular reflex. Eye-Head Coordination Gaze saccades (coordinated eye and head movements) could be controlled by a single gaze controller, with eye and head movements sharing a common motor drive. Alternatively, gaze could be controlled by coupling together separate controllers for the eye and the head. However, experimental evidence has shown that eye and head trajectories can be decoupled by changes in initial eye and head position. This evidence has led to us to propose a new model of gaze saccades based on two separate controllers, one for gaze and one for head, but without a separate controller for the eye. Clincal Eye Movement Disorders We have looked at several clinical eye movement disorders having to do with oscillations caused by aberrant firing of powerful premotor burst neurons in the brain stem. Our recent models of brain stem neurons include the biophysical properties of voltage- and ligand-dependent ion channels. This has enabled us to model many eye movement disorders showing flutter. We have also studied oculopalatal tremor (OPT). The etiology of OPT is unknown, and few treatments are successful. We have proposed a new model of hypertrophy of the inferior olive accompanied by learning in the cerebellum that reproduces most of the symptoms of OPT. It has also allowed us to recommend novel treatments to ameliorate OPT. Several patients are already being successfully treated with medications suggested by the model. We are now extending our models to account for another oscillatory eye movement that arises in some patients with multiple sclerosis. This requires a new approach to our models. In the past, we have used models of neurons that represented their state as a continuous variable (i.e., the cell's membrane potential). Now, we are using spiking neurons that communicate more like real neurons. This may be key to understanding the role of the brain's mathematical integrator in oscillatory movements. Extraocular Muscles A major concern of the NM section is to understand how the brain coordinates the movements the two eyes. Failure of this coordination can result in misalignment of the eyes, called strabismus. Strabismus occurs in about 2-6% of the population, with about half of cases occurring in childhood. Strabismus is commonly treated surgically. Over a million strabismus surgeries are performed each year in the USA, and about the same number in the EU. Surgery has a success rate that depends upon the type of strabismus, ranging from 50-80%. The short-term goal of this project is to provide a better understanding of the etiology and treatment of strabismus. Eventually this will lead to the development of improved diagnostic tests, with the hope of improving the success rate of surgery and thus reducing the number of re-operations. That will require major progress on two subprojects, modeling the orbit and modeling the brain. We are advancing our understanding of the tissues in the orbit that make eye movements. In 2003 we developed the first new model of the human orbit in 28 years. This was also the first dynamic model of the orbit. The new model is being tested on data from clinical cases of treatment for strabismus, and on data from monkeys with experimental fourth cranial nerve (IV CN) palsies. It has already uncovered some serious problems with earlier models, and shown the need for specific experimental data. This new model will provide a more physiological basis for understanding strabismus that should significantly improve the accuracy of clinical diagnoses and treatment planning. Our orbit model highlighted the incompatibility of the reported viscoelastic properties of passive extraocular muscles (EOMs) with the dynamics of eye movements in monkeys with IV CN palsies. Accordingly, we planned and executed an extensive series of experiments to measure these forces in monkeys. We found that indeed previous reports were incorrect, not only quantitatively, but also qualitatively. In particular, we discovered a wide range of non-linear properties. Interestingly, if a series of elongations is applied, the force generated by the passive muscles is only a function of the last elongation. This unsuspected feature considerably simplifies the motor control problem. To investigate how and to what extent these features also apply to activated muscles, we have now developed an experimental rabbit model that will allow us to carry out the same experiments in both passive and artificially activated muscles. So far we have already replicated the experiments on passive muscles, and found that they behave like monkey EOMs. We are also attempting to join together the anatomical, neurophysiological, neurochemical, biophysical and genetic data about how the brain moves the eyes. The model is also being generalized to include neural control of head movement, to account for control of gaze (eye in space), the real-world function of the brain. Saccade Sequences The human retina has a wide field of view, but only limited resolution. High visual acuity is restricted to a tiny region (about 1 degree) called the fovea. To see well enough to read, thread a needle, or identify faces we must point our eyes so that the image of an object of interest falls on the fovea. When information about one part of an image is acquired this way, the eye must move again to the next feature of interest. During our waking hours, people make about three saccades per second. An important question in vision processing is how the brain plans these movements. The brain could plan one movement, gather visual information, and then plan another movement. However, this would slow down the process of scanning complicated scenes. In fact, saccadic eye movements are usually planned in sequences of two or three movements. However, in the laboratory saccades are overwhelmingly studied in isolation, i.e., one at a time in response to external cues. To investigate how individual movements in a sequence are generated, we have developed a variant of the classic double-saccade paradigm. In this paradigm two targets are flashed in sequence, and the subject then makes a sequence of two saccades towards the remembered location of the flashed targets. It has been generally assumed that the subject memorizes the location of the two targets. After making a saccade toward the first one, the memorized location of the second target is then updated; the second movement is then planned and the appropriate motor command is generated. However, by tricking five human subjects into making a saccade toward the final target that is shorter than the one required to acquire the target, we found that sequences of eye movements can be generated in a radically different manner. Our results are compatible with the novel hypothesis that subjects initially compute the two motor commands required to foveate each target directly, and update the second motor command after the first movement has been executed. What is being stored are thus motor commands, and not target locations. While in the past it had been proposed that two saccadic eye movements can be planned and (partially) executed in parallel, here for the first time we show that the saccadic system embeds a motor memory that can store more than one movement command, update them, and execute them at the desired time.